Predictive model for glycemic control in patients with diabetes mellitus
![](https://ojsdergi.com/ojsicons/gor.png)
![](https://ojsdergi.com/ojsicons/pdf.png)
DOI:
https://doi.org/10.5281/zenodo.12601066Keywords:
Diabetes mellitus, glycemic control, machine learning, predictive modelingAbstract
Diabetes mellitus, a chronic disease characterized by high blood sugar levels, necessitates effective glycemic control to prevent severe complications such as damage to the heart, blood vessels, eyes, kidneys, and nerves. This study aims to utilize machine learning techniques to predict glycemic control among a open Access dataset of 77,723 newly diagnosed diabetic patients in Istanbul. By employing a logistic regression model, the study identifies key features influencing glycemic control, enhancing model interpretability for clinicians. The model demonstrates robust performance with an accuracy of 0.825, precision scores of 0.86 (positive class) and 0.76 (negative class), recall values of 0.86 (positive class) and 0.77 (negative class), and corresponding F1 scores. Feature importance analysis reveals HbA1c as the dominant predictor, significantly surpassing other variables. These findings provide critical insights into the application of machine learning in diabetes management, highlighting the pivotal role of HbA1c in glycemic control prediction.
Downloads
Published
How to Cite
Issue
Section
License
Copyright (c) 2024 Journal of Exercise Science & Physical Activity Reviews
![Creative Commons License](http://i.creativecommons.org/l/by-nc/4.0/88x31.png)
This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License.